predictores de desroden de estres

8/19/2019 predictores de desroden de estres

1/81

Studies and

Research Projects

REPORT R-786

André Marchand

Richard Boyer

Céline Nadeau

Mélissa Martin

Predictors of Posttraumatic Stress Disorders

in Police Officers

Prospective Study

Special Projects


2/81

The Institut de recherche Robert-Sauvé en santé et en sécurité du travail (IRSST),

established in Québec since 1980, is a scientific research organization well-known

for the quality of its work and the expertise of its personnel.

MissionTo contribute, through research, to the prevention

of industrial accidents and occupational diseases and

to the rehabilitation of affected workers;

To disseminate knowledge and serve as a scientific

reference centre and expert;

To provide the laboratory services and expertise

required to support the public occupational health

and safety network.

Funded by the Commission de la santé et de la sécurité

du travail, the IRSST has a board of directors made up

of an equal number of employer and worker representatives.

To find out moreVisit our Web site for complete up-to-date information

about the IRSST. All our publications

can be downloaded at no charge.

www.irsst.qc.ca

To obtain the latest information on the research carried

out or funded by the IRSST, subscribe to Prévention au

travail, the free magazine published jointly by the

IRSST and the CSST.

Subscription : www.csst.qc.ca/AbonnementPAT

Legal Deposit

Bibliothèque et Archives nationales du Québec

2013

ISBN: 978-2-89631-681-6 (PDF)ISSN: 0820-8395

IRSST – Communications and Knowledge

Transfer Division

505 De Maisonneuve Blvd. West

Montréal, Québec

H3A 3C2Phone: 514 288-1551

Fax: 514 288-7636

[email protected]

www.irsst.qc.ca

© Institut de recherche Robert-Sauvéen santé et en sécurité du travail,

July 2013

OUR RESEARCH

is working for you !


3/81

A PDF version of this publication

is available on the IRSST Web site.

Studies and

Research Projects

Predictors of Posttraumatic Stress Disordersin Police Officers

Prospective Study

Special Projects

This study was funded by the IRSST. The conclusions and recommendations are solely those of the authors.

This publication is a translation of the French original; only the original version (R-710) is authoritative.

REPORT R-786

Disclaimer

The IRSST makes no guarantee as to

the accuracy, reliability or completeness

of the information in this document.

Under no circumstances may the IRSSTbe held liable for any physical or

psychological injury or material damage

resulting from the use of this information.

Document content is protected

by Canadian intellectual property

legislation.

André Marchand1 3, Richard Boyer 2 3,

Céline Nadeau1, Mélissa Martin1

1Université du Québec à Montréal2Université de Montréal

3Centre de recherche Fernand-Seguin,

Hôpital Louis-H. Lafontaine

http://www.irsst.qc.ca/-publication-irsst-facteurs-previsionnels-developpement-etat-stress-post-traumatique-suite-evenement-traumatique-policiers-volet-prospectif-r-710.htmlhttp://www.irsst.qc.ca/-publication-irsst-facteurs-previsionnels-developpement-etat-stress-post-traumatique-suite-evenement-traumatique-policiers-volet-prospectif-r-710.html


4/81

PEER REVIEW

In compliance with IRSST policy, the research results

published in this document have been peer-reviewed.


5/81

IRSST - Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study i

Acknowledgments

The authors wish to thank all those who took part in the study, as well as the following peopleand organizations:

Research Partners

Normand Martin, Ph.D., psychologist, section chief, City of Montreal police assistance program

Longueuil police department

Saint-Jean- sur-Richelieu police department

Régie intermunicipale de police de la Rivière-du-Nord

Charles Plante, advisor, Association paritaire pour la santé et la sécurité du travail, municipal

affairs sector (APSAM)

Charles Gagné, knowledge translation advisor, Knowledge Translation and Relations withIRSST Partners

Advisory Committee

Normand Martin, Ph.D., psychologist, section chief, City of Montreal police assistance program

Yves Francoeur, president, Fraternité des policiers et policières

Mario Lanoie, vice-president, Fraternité des policiers et policières, Recherche et communications

Robert Boulé, vice-president, Fraternité des policiers et policières, Prévention et relations avecles membres

Benoit Traversy, commander, Administrative Support

François Landry, section chief, Human Resources

Consultant

Stéphane Guay, Ph.D.

Statistical Analysis

Dominic Beaulieu-Prévost, Ph.D.


6/81

ii Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study - IRSST

Research Assistants

Sophie Lacerte, B.Sc.

Myra Gravel Crevier, B.Sc.

Kim Regaudie, B.Sc.

Reviewers

Mélissa Martin, Ph.D.

Joannie Poirier-Bisson, B.Sc.

Roseline Massicotte, M.Ps.

Marie-Josée Lessard, M.Ps.

Myra Gravel Crevier, B.Sc.

Mariko Chartier-Otis, M.Ps.

Andréa Leduc, M.Ps.

Sandra Primiano, M.Ps.

Guillaume Foldes-Busque, Ph.D.


7/81

IRSST - Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study iii

Abstract

Occupational health and safety and specific objectives of project – The purpose of thisresearch project was to arrive at a better understanding of the development of posttraumaticstress disorder (PTSD) following a work-related accident by studying the associated risk and

protective factors. The project originated with a request from the police, as officers are oftenexposed to traumatic events (TEs). The researchers worked on the hypothesis that peri- and posttraumatic factors could better explain the development of PTSD and workers’ ability to copewith TEs than pretraumatic factors. This study is original in that it focuses on little-researched protective factors, distinguishes three types of factors (pre-, peri- and posttraumatic) andconcerns Quebec police officers. It is innovative in that it includes both men and women and hasnot only a retrospective component, but also an additional prospective analysis essential to a better understanding of the predictors of PTSD. While this research report concentrates on thefindings of the prospective analysis, it also makes connections with the data presented in theretrospective research report [1].

Method – Eighty-three officers of the Service de Police de la Ville de Montréal (SPVM) andother police departments in Quebec participated voluntarily in a repeated-measures prospective study (quasi-experimental design). They had all been involved in a major traumatic event between May 2006 and May 2010. They were assessed, on average, between 5 and 15 days, 1month, 3 months and 12 months after the event. The same measurement instruments as for theretrospective study — semistructured interviews and self-reports—were used to determinewhether or not the police officer was suffering from PTSD and to evaluate various predictorsassociated with PTSD development. These instruments were chosen for their psychometric andclinical properties and because they offer appropriate testing of the research hypotheses.Multivariate statistical analysis was conducted to identify more specifically the main predictorsat work and the strength of their impact on PTSD.

Findings – The prospective study findings show that 3% of police officers suffered from clinicalPTSD, while 9% experienced partial PTSD. In contrast, data from the retrospective study showthat 7.6% of the officers in the sample had clinical PTSD, whereas 6.8% were affected by partialPTSD. In the prospective component, the regression analysis results indicate that posttraumaticrisk factors (i.e., the symptoms of acute stress disorder [ASD] and depression) are the main predictors. Pretraumatic (emotional stress-management techniques) and peritraumatic(peritraumatic distress and dissociation) risk factors, though less critical, are still significant. Wedid not observe any protective factors associated negatively with PTSD symptoms. The results ofmultiple logistic regression analysis from the retrospective study suggest that peritraumatic riskand protective factors (i.e., dissociation and social support during the event) are the chief predictors. The results of the retrospective study descriptive analyses show that officers resorted

to a variety of ways and strategies to cope with a critical event at work. The officers said thatwhat particularly helped them after a TE was to talk to colleagues about it, make use of supportservices and have time off. Their advice to other officers who are exposed to a critical event is totalk about it and to see a psychologist, and the majority of them are open to receiving this kind ofsupport service if needed.


8/81

iv Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study - IRSST

Conclusions – The low rate of PTSD among police officers assessed in the study contrasts withinitial expectations. It suggests that they are resilient, despite the fact that they represent a population with a high risk of experiencing TEs in the course of their work. The study findingscorroborate a number of elements found in the literature on various populations, including the police. Since the factors associated with the development of PTSD in police officers

(i.e., dissociation, emotional and physical reactions, ASD, symptoms of depression, emotionalstress-management techniques) can potentially be mitigated or prevented, specific, tailoredmeasures could be developed to target these factors with a view to achieving better results in the prevention of PTSD. The factors associated with coping following trauma (i.e., hardiness, socialsupport) can be developed or strengthened through preventive strategies that should generally beincorporated into police training programs. The findings of this study could enhance traininggiven as part of the police officer assistance program (PAPP) of the SPVM and other policedepartments. They also underscore the importance of taking a preventive approach, which isalready the case at the SPVM. The approach could also be adopted by the employee assistance programs of other police forces.

Foreseeable impact – This study, the first of its kind in Quebec, could serve as a benchmark forfurther research using a sample of Quebec police officers. The knowledge gained should helpscreen for and prevent PTSD. In addition, police departments should be able to use the studyrecommendations to develop strategies that foster the development of TE protective mechanismsand reduction of the risk factors involved. The study could have a significant impact on otheroccupational groups at high risk of exposure to TEs, such as firefighters, ambulance attendants,first responders and social workers.


9/81

IRSST - Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study v

Contents

1. INTRODUCTION ...................................................................................................... 1

2. BACKGROUND ....................................................................................................... 2

2.1 Pretraumatic Risk Factors ............................................................................................... 3 2.1.1 In General Population ..................................................................................................... 3 2.1.2 In Police Population ........................................................................................................ 3

2.2 Peritraumatic Risk Factors .............................................................................................. 4

2.2.1

In General Population ..................................................................................................... 4

2.2.2

In Police Population ........................................................................................................ 5

2.3 Posttraumatic Risk Factors .............................................................................................. 5 2.3.1 In General Population ..................................................................................................... 5 2.3.2 In Police Population ........................................................................................................ 5

2.4 Pretraumatic Protective factors....................................................................................... 6

2.4.1 In General Population ..................................................................................................... 6 2.4.2

In Police Population ........................................................................................................ 6

2.5

Peritraumatic Protective Factors .................................................................................... 7

2.6 Posttraumatic Protective Factors .................................................................................... 7

2.7 Influence of Various Predictors in Population ............................................................... 7

2.8 Limitations and Scope of Current Knowledge ............................................................... 8

2.9 General Purpose of Research ........................................................................................... 9

2.10 Research Hypotheses ...................................................................................................... 10

3.

RESEARCH METHODS ........................................................................................ 11

3.1 Procedure ......................................................................................................................... 11 3.1.1

Police Officers Involved in TEs .................................................................................... 12

3.2 Measurement Instruments Used .................................................................................... 13 3.2.1 Diagnostic Measurement of PTSD and Other Mental Disorders.................................. 14 3.2.2 Measurement of Pretraumatic Factors .......................................................................... 14


10/81

vi Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study - IRSST

3.2.3 Measurement of Peritraumatic Factors ......................................................................... 16 3.2.4 Measurement of Posttraumatic Factors ......................................................................... 17

3.3 Study Design and Variables ........................................................................................... 23

3.4 Sample Size ...................................................................................................................... 23

3.5 Ethical Considerations.................................................................................................... 24

4.

FINDINGS .............................................................................................................. 25

4.1 Participation Rate ........................................................................................................... 25

4.2 Quantitative Data ............................................................................................................ 26 4.2.1

Statistical Analysis ........................................................................................................ 26

4.2.2 Description of Officers Involved in TEs ....................................................................... 26 4.2.3 Description of TE .......................................................................................................... 27

4.2.4

Prevalence of ASD and PTSD ...................................................................................... 29

4.3 Predictive Data ................................................................................................................ 30 4.3.1

Prediction of PTSD Symptoms ..................................................................................... 30

4.3.1.1 Missing-Data Handling Procedure ........................................................................ 30 4.3.1.2 Statistical Analysis Models ................................................................................... 31 4.3.1.3

Interpretation of PTSD Symptom Predictive Models at T2, T3 and T4 ............... 35

5. DISCUSSION ......................................................................................................... 37

5.1 Quantitative Data ............................................................................................................ 37

5.2 Predictive Data ................................................................................................................ 39

5.2.1 PTSD Predictors............................................................................................................ 39 5.2.2

Peritraumatic Distress and Dissociation ....................................................................... 42

5.2.3 Social Support ............................................................................................................... 43 5.2.4 Depression, ASD and Emotion Management ............................................................... 44

5.3 Clinical Implications ....................................................................................................... 45

5.3.1 Primary Prevention ....................................................................................................... 45 5.3.2

Secondary Prevention ................................................................................................... 46

5.3.3 Evaluation ..................................................................................................................... 46

5.3.4

Tertiary Prevention ....................................................................................................... 47

5.4 Limitations and Scope of Findings ................................................................................ 48

5.5 Future Research .............................................................................................................. 49

6. CONCLUSION ....................................................................................................... 52


11/81

IRSST - Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study vii

7. APPLICABILITY OF FINDINGS ............................................................................ 53

8. POTENTIAL IMPACT ............................................................................................ 53

9. LIST OF SCIENTIFIC PUBLICATIONS PRODUCED WITH THIS FUNDING ....... 54

10. BIBLIOGRAPHY ................................................................................................ 58


12/81

viii Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study - IRSST

Tables

Table 1 Sociodemographic characteristics of police officers exposed to a

traumatic event ( N = 83) ......................................................................................13

Table 2 Main variables measured at different assessment times ..................................20Table 3 Research protocol for repeated-measures prospective study ..........................23

Table 4 ASD and PTSD diagnoses of participants, at each time of

assessment .............................................................................................................30

Table 5 Description of variables statistically correlated to MPSS scores at

T2, T3 and T4 .......................................................................................................33

Table 6 Predictive model for PTSD symptoms at T2 ( n = 71) .......................................34Table 7 Predictive model for PTSD symptoms at T3 ( n = 69) .......................................34Table 8 Predictive model for PTSD symptoms at T4 ( n = 72) .......................................35Table 9 Summary of main predictors, according to retrospective and

prospective parts of study ...................................................................................40

Figures

Figure 1 Number of participants at each time of assessment ..........................................25

Figure 2 Ranks of police officers ( N = 83), by category ...................................................27Figure 3 Characteristics of traumatic events ( N = 83) .....................................................28Figure 4 Emotions experienced during or immediately after TE ( N = 83) ....................29


13/81

IRSST - Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study 1

1. INTRODUCTION

Police officers, due to the nature of their work, are continually exposed to incidents characterized by unpredictability, danger to their personal safety and close contact with death (e.g., trafficaccidents, shootings, suicides, homicides). An analysis of data from Quebec’s Commission de la

santé et de la sécurité du travail (CSST)1

shows that police officers are one of the occupationalgroups most often exposed to violence in the workplace [2]. Workplace violence and traumaticevents (TE) can have a profound impact on a person’s psychosocial functioning and even lead to posttraumatic stress disorder (PTSD) [3-5]. Significant posttraumatic stress reactions are seen inclose to 30% of women and slightly over 15% of men who are victims of violence at work [2].These figures indicate that posttraumatic stress is the most frequent effect for all workers. As thesame authors note [2], research in the field must aim to reduce the risks and consequences of this problem in the occupational groups most affected, which include police officers.

The purpose of this study, which originated with a request for assistance from the police officerassistance program (PAPP) of the Service de Police de la Ville de Montréal (SPVM) 2 and the

SPVM labour-management joint committee, was to identify the predictors of PTSD to promotemore effective prevention. The study reflects the objectives of the Association paritaire pour lasanté et la sécurité du travail, secteur “affaires municipales” (APSAM). 3 This labour-management health and safety organization, which seeks to develop ways to protect health,ensure safety and physical integrity and foster greater responsibility by the police community, is becoming increasingly aware of the psychosocial risks of police work. Indeed, it is determined totackle the issue of the trauma to which police officers are exposed in performing their duties [6].As a result, APSAM also lent its support to the research.

Before we go any further, it is important to explain that PTSD is a reaction to a traumatic event(a stressor). The victim has experienced, witnessed or been confronted with an event or events thatinvolve actual or threatened death or serious injury, or a threat to the physical integrity of oneselfor others. In addition, the person’s response generally involves intense fear, helplessness orhorror [7]. Police officers work in hazardous environments in which they are liable to experienceTEs.4 At the SPVM, approximately 9 major events (i.e., an event involving at least four officers)and over 50 traumatic individual and specific critical incidents are reported every year (SPVM joint committee, personal communication, June 6, 2003).

1 CSST: Quebec occupational health and safety board.

2 SPVM: City of Montreal police department.

3 APSAM: Joint labour-management association for occupational health and safety, municipal affairs sector.

4 The term “traumatic event” (TE) is used throughout this report to refer to any event with the potential totraumatize. A TE does not necessarily lead to posttraumatic reactions. A critical incident or major event, to use

police jargon, is considered to be an event that could potentially be traumatic.


14/81

2 Predictors of Posttraumatic Stress Disorder in Police Officers: Prospective Study - IRSST

2. BACKGROUND

The lifetime prevalence rate of PTSD for groups with a high risk of experiencing a TE isgenerally estimated to be between 30% and 45% [8], while for the general population, it is between 7% and 8% [9, 10]. Police officers are a high-risk population for exposure to TEs and

subsequent development of PTSD. While some studies report higher rates of PTSD prevalencefor police than for the general population (13%–35%; [11-14]), others note similar prevalencerates for the two populations (6%–9%; [15-19]). Some police forces do not appear to be anymore at risk of developing PTSD than the general population, even if their officers are morefrequently exposed to TEs. None of these studies evaluated Canadian or Quebec police officersspecifically. As a result, there is no data at present on PTSD prevalence in the Canadian orQuebec police populations. However, of CSST-compensated injuries between 2000 and 2002,the number of posttraumatic stress cases was four times higher for police officers than for allQuebec workers taken together [20].

PTSD causes persistent avoidance behaviour with respect to stimuli associated with the event, as

well as a general numbing of reactions, sustained symptoms of neurovegetative activation,constant reliving of the traumatic event and a significant deterioration in ability to function(i.e., problems at work and in interpersonal relationships) [7]. Psychologically, the sufferingcaused by PTSD can be so severe that it can trigger anxiety disorder, depression, psychotropicdrug abuse, divorce and even suicide [21, 22]. Physically, the impact of a traumatic experiencecan result in increased use of health care services and medication. At work, this psychologicaldisorder can have far-reaching consequences, with exorbitant costs for the individual and theorganization: reduced performance at work and absenteeism, sharp increase in compensation payments and insurance claims to the CSST or other government agencies [23, 24].

While experiencing a TE is a necessary condition for developing PTSD, it is not a sufficientcondition. Whether or not an individual can resume a normal life after a TE often depends on the presence or absence of various factors that influence posttraumatic reactions. For instance, riskfactors increase the probability that exposure to a TE will have a long-term impact on theindividual’s psychological well-being. They are associated with the development, maintenanceor exacerbation of posttraumatic symptoms. In contrast, protective factors make coping easierfollowing a TE by preventing posttraumatic symptoms or reducing their intensity [25]. Moreextensive knowledge of the various predictors of posttraumatic symptoms will help preventdevelopment of PTSD in police officers and provide more appropriate treatment to those whohave it. Preventing PTSD in police officers is essential, as they are constantly exposed to TEsthat have the potential to cause posttraumatic reactions [26].

This research project, which is the second, prospective, part of the study, seeks to clearly identify

the risk and protective factors that influence the development of PTSD following a traumaticevent, with a view to preventing the disorder in a police population. Current knowledge aboutPTSD predictors is set out below. The factors involved can be divided into three distinctcategories: pretraumatic factors (characteristics that existed before the trauma), peritraumatic(specific circumstances at time of and immediately after the TE) and posttraumatic(characteristics that develop following the trauma). While the list of factors is not exhaustive, themost relevant predictors, and those for which there is significant empirical evidence of theirinvolvement in posttraumatic reactions, are examined briefly in the following sections. A more


15/81


comprehensive review of risk and protective factors is provided in the book chapter by Martin,Germain and Marchand [27].

2.1 Pretraumatic Risk Factors

2.1.1

In General Population

In various population groups, personal and family psychiatric and psychological histories(e.g., substance abuse, dependency, depression, anxiety) or past trauma (e.g., physical or sexualabuse) are risk factors for the development of PTSD [28-31]. As well, having strong fundamental beliefs that the world is good, just and logical and believing in one’s worth and personalinvulnerability could make a person more susceptible to increased pathological symptomsfollowing a TE [32]. Having such beliefs proven unfounded by a traumatic experience andencountering problems coming to terms with the experience raise the probability that the victimwill develop and continue to suffer from PTSD symptoms [32].

Women appear to be more vulnerable than men to developing PTSD [30, 33, 34] and to sufferingfrom it as a chronic condition [31]. However, the TEs experienced by women differ from thosesuffered by men. Women are more likely to be victims of sexual assault, whereas men are moreoften the targets of physical assaults [9, 33]. The event itself, rather than gender, may beassociated with greater vulnerability. However, even when the type of trauma experienced iscontrolled for, women seem to be more vulnerable than men to the development of PTSD [9,35]. In contrast, other studies have not found any association between gender and PTSD [36-38].So far, a range of sociodemographic data (e.g., age, gender, ethnic group, level of education)have been studied as possible predictors. The findings of these studies have not been conclusive.This would seem to be due, in part, to the fact that sociodemographic data often involveconcomitant or confounding variables that can themselves influence or even explain people’s

vulnerability or resilience. To take just one example, a low level of education may be associatedwith other variables, such as low socioeconomic status. It is therefore not easy to determine the predictive force of sociodemographic data, as findings vary from one study to the next [27].

2.1.2 In Police Population

Among police officers, greater exposure to TEs at work over the last year, along with anaccumulation of traumatic experiences or stressors in the workplace, predisposes them to thedevelopment of posttraumatic reactions [14, 39-42]. Other studies have shown that for youngofficers, less experience increases the probability that they will develop PTSD symptomsfollowing a TE. This is likely due to the fact that young officers have not developed sufficient

coping strategies to deal with the high level of stress associated with police work [17, 43]. Theimpact of past personal traumas (e.g., during childhood) also seems to be a significant factor in predicting short- and long-term posttraumatic symptoms in police [44, 45]. Some authors havenoted, in a prospective study of police officers, that family psychopathological history (moodand anxiety disorders and substance abuse) is a vulnerability factor in the development of PTSDfollowing a traumatic experience [46]. In addition, police officers who reported higher levels oforganizational stress had more serious posttraumatic symptoms at the time of the study [40, 47-49]. When it comes to sociodemographic factors such as age, gender, social standing, marital


16/81


status and level of education, the findings of various studies are mixed. However, being ofHispanic origin is a risk factor associated with more serious posttraumatic symptoms, even aftercontrolling for other sociodemographic data and the frequency and severity of exposure tocritical incidents [50, 51]. Contrary to the general population, women police officers do notappear to be any more at risk of developing PTSD than their male coworkers [13, 43, 50, 52].

Neurotic personality traits, involving sensitivity to negative stimuli, and dissociative traits, foundin people who tend to experience a state of dissociation when exposed to a TE, are associatedwith more-intense PTSD symptoms among police officers [17, 49, 53, 54]. Moreover, having anexternal locus of control (i.e., constantly perceiving events to be caused by external forces beyond one’s control) seems to be related to the development of posttraumatic symptoms. Policeofficers who repress their emotions, have trouble expressing them and those who tend to showtheir anger more have a higher risk of developing posttraumatic stress reactions [15, 55].Sensitivity to anxiety (i.e., being afraid of the consequences associated with the physicalsymptoms of anxiety) is also a PTSD risk factor for police [12]. Lastly, from a psychobiologicalstandpoint, one study has suggested that hypersensitivity to threatening situations, high reactivityof the sympathetic nervous system when confronted with an explicit, imminent threat, and a

failure to cope with repeated aversive stimuli (slower habituation) are prospective predictors ofthe severity of posttraumatic symptoms in novice police officers [56].

2.2 Peritraumatic Risk Factors

2.2.1 In General Population

In some population groups, those who experience strong negative emotional reactions (e.g., fear,guilt, shame, anger, disgust, sadness) or strong physical reactions of anxiety (e.g., palpitations,trembling, dizziness, sweating, hot flushes or shivering) during and immediately after the TE aremore vulnerable to the development of PTSD [57-60]. More recently, researchers have been

increasingly focusing on a more global construct, called peritraumatic distress. It refers to thenegative emotional responses and physical reactions of anxiety experienced during orimmediately after a TE [61]. Among the victims of an industrial disaster, peritraumatic distresswas found to be a predictor of PTSD [62]. One of the key components of peritraumatic distress isthe notion of loss of control (i.e., of one’s emotions, of one’s sphincter, bladder or bowels),which is directly associated with the development of posttraumatic symptomatology [63]. Inaddition, many studies suggest that peritraumatic dissociation (i.e., detachment, lack ofemotional reactivity, sense of derealization, depersonalization, etc.) is a vulnerability factor inthe emergence of posttraumatic stress symptoms [57, 58, 62, 64-66]. Dissociative reactionsgenerally occur in a situation where peritraumatic distress is high, although some people canexperience intense distress without dissociating [61, 67]. Lastly, the meta-analyses conducted by

Ozer et al. [68], Breh and Seidler [69] and Lensvelt-Mulders et al. [70], which cover,respectively, 35, 59 and 68 empirical studies, confirm that peritraumatic dissociation is a highrisk factor in the development of PTSD.

A study of traffic accident victims shows that an initial feeling of fright at the time of the eventsubstantially increases the risk of developing PTSD [71]. Fright is a reaction that shares some ofthe characteristics of peritraumatic dissociation. In fact, fright essentially involves a completeabsence of emotion, of thought or words, or the feeling of being frozen during part of the event.


17/81


Another study, of patients who had suffered a heart attack, shows that fright is associated with posttraumatic symptoms three months after the event [72]. TE severity has also been associatedwith the development of PTSD [37, 59, 73, 74]. On the other hand, the findings of the variousstudies are mixed, as in a study of survivors, the severity of the TE was not a predictor of posttraumatic symptoms [75]. Some researchers measure the severity of the TE by considering

the objective characteristics associated with the event. Of these characteristics, the duration ofthe event, for instance, influences the posttraumatic stress reactions [76, 77]. Thus, the longer theexposure to the TE, the more severe the PTSD symptoms [78]. At the same time, incidentsinvolving the discovery of injured and deceased people are more likely to cause the developmentof PTSD [79], just as, in clinical experience, unexpected or deliberate events that put theindividual’s life in danger or that are physically or psychologically degrading are likely to.Events that result in bodily injuries are associated more with the development of PTSD thantraumatic experiences that do not cause such injuries [80, 81].


Among police officers, peritraumatic distress, peritraumatic dissociation and emotional and physical reactions during trauma are predictors of PTSD [15, 17, 18, 43, 46, 50, 51, 53, 54, 61,82]. A number of studies have shown that exposure to death during the TE, the existence of athreat to human life or to the personal safety of the officer or the officer’s partner, and theseverity of the exposure are risk factors in the development of PTSD [15, 17, 19, 43, 82-84].However, one study indicates that the perception of proximity to death is not associated with thedevelopment of PTSD [85]. Certain types of TE—such as witnessing gunshots that put one’sown life or a coworker’s life in danger, witnessing child abuse, death, including homicidevictims, or victims of serious traffic accidents—seem to have a major impact on the developmentof posttraumatic symptoms in the police population [42, 44]. Lastly, one study has noted that police officers who perform what are for them unusual tasks during major events (e.g. doing

what a firefighter would usually do) have a higher risk of developing PTSD than those who carryout their usual police duties [19].

2.3 Posttraumatic Risk Factors


Among accident victims, depressive symptoms [75], additional stressors that arise following theTE (e.g., job loss, financial or legal problems, illness or death of a family member or friend) [86,87] and acute stress disorder (ASD) [62, 88-90] are indicators for the future development ofPTSD. Ingram et al. [91] have shown that a social network that offers little or no support is

associated with poor posttraumatic adjustment.


Among police officers, the limited amount of time granted by the employer to recover from theTE, the general dissatisfaction with the organizational support offered and the lack of supportoutside of work after the TE seem to be factors that predict the development of posttraumaticsymptoms [15]. Physical injury from the TE, depressive symptoms and post-TE negative life


18/81


events appear to be posttraumatic risk factors in the development of PTSD in police officers [12,48, 54, 85]. Officers who report having used avoidance-based stress-management techniques tocope with the TE tend to experience greater posttraumatic symptomatology [17, 49, 51, 92, 93].In addition, among police officers, psychological debriefing after the event either has no effecton PTSD symptoms [94] or may cause more symptoms [95, 96].

2.4 Pretraumatic Protective factors


People can use a variety of behavioural or cognitive stress management strategies [97]. Amongthe usual stress management strategies, there are problem-focused coping strategies (effortsaimed to change the situation by using a problem-solving strategy) and emotion-focusedstrategies (efforts aimed to change the negative interpretation of the situation and to soothenegative emotions through strategies such as avoidance or daydreaming) [97]. Individuals tend touse the same stress management method throughout their lives, which means that the use of

stress management strategies is considered to be a relatively stable pretraumatic characteristic.Persian Gulf War returnees who did not have PTSD symptoms tended to use more problem-focused coping strategies and fewer strategies based on avoidance and daydreaming than theircolleagues with PTSD [98]. A stress-resilient or stress-resistant personality copes better after aTE [99]. In fact, an individual with that kind of hardiness assesses stressful events appropriatelyand shows an ability to cope with them when they arise [100]. More specifically, this personalitytype is characterized by an aptitude for engagement in the community and a strong feeling ofcontrol that enables the person to make decisions and assume responsibility for them. It is alsorelated to flexibility, optimism, good self-esteem, ability to face challenges, the capacity toattribute events appropriately and to make sense of events experienced [101, 102]. Furthermore,a feeling of self-efficacy (i.e., having high expectations of one’s own effectiveness in performing

various activities and solving problems) is associated with less psychological distress and fewer posttraumatic symptoms in victims of natural disasters [103] and firefighters [104, 105]. Peoplewith a high sense of self-efficacy expect to be successful, solve problems, overcome challengesand be able to handle stressful situations.


Some studies have found that past experience and appropriate training—that is, having facedstressful situations in the past or having being trained to know how to react to them—are protective factors that help police officers to deal with posttraumatic stress reactions [17, 19,106, 107]. For individuals at high risk of being involved in a TE, experience may help them feel

better equipped to handle what may arise, more in control and less stressed by their work.Appropriate training helps officers to understand and feel comfortable in a well-defined role,reducing their uncertainty about what to do in a given situation and making them more efficientat work [108]. However, other studies of police officers have found that experience and trainingare not significant protective factors [15, 109]. Furthermore, hardiness, especially theengagement aspect, has been associated with fewer PTSD symptoms among female officers [11,18, 57]. The findings of the various studies that have examined the use of stress-management


19/81


strategies (emotion-focused coping and problem-focused coping) have been inconclusive aboutwhich one might be beneficial to police officers who have been exposed to a TE [17, 92, 110].

2.5 Peritraumatic Protective Factors

To our knowledge, there are no documented peritraumatic protective factors and predictors forPTSD in various populations. However, the first, retrospective, part of this research projectrevealed a new peritraumatic protective factor in the scientific literature on the police population: peritraumatic social support. It has been shown that officers who benefit from the support of theircoworkers during or immediately after a TE are less at risk of developing PTSD symptoms [18].

2.6 Posttraumatic Protective Factors

In a number of different populations, appropriate or positive social support after the TE seems toreduce the effects of the stress [111-114]. Family, friends, partner and coworkers are mentionedas major sources of support [101]. In addition, a psychological debriefing, which is a short, early

intervention immediately following a TE, can also be effective in supporting the individualsinvolved in these incidents and, in some cases, can help prevent the development of PTSD [26].However, these findings are increasingly being contested in the scientific community. Indeed,more and more guidelines, including those of the National Institute for Clinical Excellence(NICE) [115], the Australian Centre for Posttraumatic Mental Health (ACPMH) [116] and theAmerican Psychiatric Association (APA) [117], do not recommend the use of debriefing.

Among police officers, those who perceive greater availability of social support [40, 51, 82],greater satisfaction with the support received [85] and greater emotional support from coworkersand supervisors have fewer posttraumatic symptoms. Officers who perceive a certain opennessfrom their coworkers when they express their emotions and those who feel at ease talking about

their traumatic experiences and the emotional impact caused by them in the workplaceexperience fewer PTSD symptoms [94, 118]. Moreover, positive social support from coworkers[18] and supervisors is reportedly one of the types of support that has the most impact among police officers [52]. Being able to make sense of the traumatic event and feeling that one has theresources required to deal with it—which is also known as a sense of coherence—help reducePTSD symptoms in police officers [40]. Lastly, providing early intervention, such as psychological debriefing, to police officers after a TE has been shown to have no effect on posttraumatic symptoms [94] and may even give rise to more symptoms [15, 95, 96].

2.7 Influence of Various Predictors in Population

Two recent meta-analyses of PTSD predictors in different populations (civilian and military)highlight interesting findings about the influence of these predictors on the development ofPTSD. Brewin et al.’s meta-analysis [86] examines 14 risk factors considered in 77 studies andnotes that the most significant risk factors for the subsequent development of PTSD are peritraumatic factors (e.g., severity of TE) and posttraumatic ones (e.g., inadequate socialsupport and subsequent stressful events), as opposed to pretraumatic factors. Ozer et al. [68] alsosuggest, in their meta-analysis of 7 predictors from 68 studies, that peritraumatic (e.g., perceptionof threat to life, dissociation and emotional reactions) and posttraumatic factors (e.g., perceived


20/81


social support) are more important than pretraumatic factors in explaining the origin of PTSD.Overall, the two meta-analyses reveal findings that are more or less equivalent to those obtainedfor the police population. They also indicate that the predictive power of the various predictors isnot uniform and that the extent of the effect of each predictor may depend on the population inquestion or other factors. For example, Brewin et al. [86] report that the following predictors are

significantly more important in military populations than among civilians: young age at time oftrauma, low level of education, minority status, trauma severity and lack of social support.

2.8 Limitations and Scope of Current Knowledge

An increasing number of studies are seeking to identify PTSD predictors more accurately inorder to gain greater insight into the development of PTSD and be able to devise preventionstrategies. This is a first step in the advancement of current knowledge. However, much of theresearch in this field suffers from significant methodological shortcomings.

For instance, most of the studies use self-reports, or instruments having restricted psychometric

properties, as well as divergent instruments, which considerably limits the interpretation ofresults and makes it difficult to compare studies. For example, some studies seek to measuresymptoms and arrive at a diagnosis of PTSD strictly on the basis of questionnaires or interviewsdesigned by the researchers. As a result, in a high number of cases, the measurement instrumentsor clinical interviews do not have the psychometric properties required to eliminate sources oferror and variations or inconsistency in results stemming from factors related to the instrumentitself. Whenever possible, it is essential to use not only the same measurement instruments fromone study to the next, but also validated, consistent questionnaires and structured interviewshaving clinical criteria that have been determined by a large number of past studies (SCID-I,CAPS, MPSS, etc.). Moreover, most studies are retrospective. They do not provide as accurateanalysis of the impact of predictors as prospective studies do because they are subject to recall bias. To address these shortcomings, researchers should consider conducting prospective andlongitudinal studies. They allow evaluation of the development of the PTSD and the mechanismsthat come into play during the period following the TE. However, these studies still need to beconducted over a sufficiently long period of time (e.g., one year or more) to ensure arepresentative sampling with respect not only to average level of exposure to stressful events, butalso to posttraumatic symptoms usually shown by police officers. Some studies measuresymptoms and reactions following a TE, but do not specifically assess the diagnosis of ASD andPTSD. Moreover, many retrospective studies assess only the current diagnosis of PTSD eventhough the event itself may have taken place several years earlier. In fact, the individual mayhave suffered from PTSD during the months following the TE, but has now recovered from it.Studies that assess only current PTSD overlook valuable information. Moreover, the risk factorsfor developing chronic PTSD (i.e., PTSD that has been present for over three months) coulddiffer substantially from those for acute, non-chronic PTSD. Rarely do studies take this fact intoconsideration, however. Furthermore, studies that assess a diagnosis of both current PTSD andlifetime PTSD often neglect to distinguish whether the PTSD diagnosis is related to the TE thatis the subject of the study, or rather whether it is related to some other trauma that occurred inchildhood or adulthood [36]. A number of other studies have also examined the associations between various factors and PTSD using simple regressions or correlations only, transversely but


21/81


not prospectively. The disadvantage of this type of analysis is that it cannot establish the causalmechanisms involved, nor identify the best predictors of PTSD.

A variety of different populations—soldiers, accident victims, disaster victims, etc.—have beenstudied so far, which has broadened our knowledge base. However, few studies have focused on populations at high risk of experiencing TEs, except for soldiers. It is clear that insufficient workhas been done on populations at high risk of developing posttraumatic reactions, such as front-line emergency workers (e.g., police officers). Risk and protective factors form a dyad and must be studied jointly in order to gain a better understanding of PTSD development. However, mostof the literature has concentrated on risk factors rather than protective factors. Moreover, onlyrarely have studies explored risk factors in a police population. Even rarer are studies that haveassessed protective factors that facilitate recovery following exposure to a TE among policeofficers. Yet learning more about these factors to prevent PTSD in a high-risk population isabsolutely essential.

To sum up, a significant amount of work has been done in the field of trauma, but the findingshave been inconclusive. There is currently no consensus about the significance and predictive power of certain factors that influence PTSD symptoms. These ambiguous findings can beexplained by the many methodological shortcomings of PTSD research, such as the lack ofrigour of many studies, interference between variables, the difficulty of isolating some variablesand the retrospective character of most studies. All these weaknesses significantly limitinterpretation of the connections found between predictors and PTSD. Further research in thisfield is required due to the lack of studies on police populations, the fact that many aspects haveyet to be explored and because of our limited knowledge on the subject. Our research is an effortin this direction and aims to learn more about protective factors. Indeed, this current project is ahigh priority, as it was initiated by a request from the PAPP and the SPVM joint committee andis supported by APSAM. Furthermore, it covers both men and women, and women are a population that has not been studied much in the literature. It will therefore be possible to

compare findings with respect to gender and determine whether there are any differences in thisregard.

2.9 General Purpose of Research

The purpose of this research project is to address the problem concerning the lack of studies with police populations, the lack of prospective studies and the methodological shortcomings thatcurrently exist in the literature, and to expand our knowledge about the predictors of PTSD in police. This study is the second, prospective, part of the research program. In addition, indiscussing the findings, this research report also seeks to establish certain relationships or drawcomparisons with the data produced by the first, retrospective, part of the research program. It

therefore addresses the overall objective of the research project, which is to gain a betterunderstanding of the development of PTSD in police officers and of the factors involved. Morespecifically, the study assesses the risk factors that increase police officers’ vulnerability todeveloping PTSD, as well as the protective factors that make it easier for them to cope followinga traumatic event.


22/81


2.10 Research Hypotheses

Based on data from the literature on various populations, police in particular, and from theretrospective part of this research project, the hypotheses of this prospective study are as follows:

1.

The development of PTSD symptoms and ability to cope following a TE can be betterexplained by peri- and posttraumatic factors than by pretraumatic ones.

2.

The development of PTSD symptoms can be explained, mainly by the following risk factors:dissociative reactions, severity of the event and perception of negative social interactions.

3. The ability to cope following a traumatic event is chiefly tied to a perception of positivesocial support. As data on protective factors among police officers are scarce, with theexception of support, the impact of the following factors will be examined on an exploratory basis: stress management strategies, hardiness, and years worked as a police officer beforethe traumatic event.

4.

On a descriptive and exploratory basis, for information purposes, we intend to track not onlythe extent of the prevalence of PTSD and changes in PTSD symptoms over time, but also thedistinctive, predictive power, if any, of the risk and protective factors that have an influenceon the development of PTSD at different times.


23/81


3. RESEARCH METHODS

3.1 Procedure

Police officers were informed of the study through articles published in police brotherhood and

police department newsletters. The project was also presented on numerous occasions to seniormanagers of the SPVM’s various divisions, police assistance program officials and the labour-management joint committee. At these meetings, we outlined the project and provided literatureand notices to be posted in workplaces, explaining the main points of the study and how tocontact the research coordinator following a traumatic event. Overall, most referrals came from police officers themselves (e.g., participants, supervisors, lieutenants and commanders), from thecorporate operations division and from the PAPP. Moreover, the fact that a number of majorevents were covered by newspapers and television enabled the research team to target certainunits specifically, to ask them to take part in the study. Several SPVM officers involved in TEs between May 2006 and May 2010 agreed to participate in the prospective study (n = 72). Anumber of officers from other police departments (e.g., Longueuil, Saint-Jean-sur-Richelieu,

Régie intermunicipale de police de la Rivière-du-Nord) were also recruited for the study (n = 11)to ensure that the sample size was large enough to test our hypotheses.

Police officers interested in taking part in the study were asked to contact the project coordinator,who conducted initial screening of participants by telephone. During the telephone interview,officers were informed about what the study entailed, the use of various measurementinstruments and data confidentiality. Officers who were still interested in taking part and whomet the study criteria were given an initial appointment. They were also urged to suggest to their police partners and other coworkers that they, too, take part in the study if they were involved inthe same TE. Participant assessments were held in places where officers felt comfortable, eitheron the campus of the Université du Québec à Montréal, at the neighbourhood police station or at

home. Before starting the assessment, participants were asked to read and sign the consent form.The form stated the purpose of the study—stress reactions following a TE—but did not mentionany research hypotheses in order to avoid raising participants’ expectations. Participants wereassessed at four different times: from 5 to 15 days after the TE (time 1), then 1 month (time 2),3 months (time 3) and 12 months (time 4) after the event.

The inclusion criteria were more or less the same as for the retrospective study. All the officershad recently been exposed to a TE as defined by PTSD diagnostic criterion A. Under thiscriterion, the person has experienced, witnessed or been confronted with an event that involvedactual or threatened death or serious injury, or a threat to the physical integrity of oneself or others.Furthermore, in order to be considered traumatic, the event must have provoked a response that

involved intense fear, helplessness or horror. If the police officer did not feel these emotionsduring the traumatic event, but reported a response of anger, guilt or shame, the research teamdeemed that the event still qualified as being traumatic. The other inclusion criteria were theability to speak and read either French or English and being capable of undergoing asemistructured clinical interview and answering questions. The exclusion criteria were a psychotic state, an organic mental disorder, signs of suicidal intent and any serious illness that, inthe evaluator’s judgment, could entail a risk for the participant. As it turned out, no participantwas excluded from the study for the above-mentioned reasons.


24/81


3.1.1 Police Officers Involved in TEs

Eighty-three police officers (63 men and 20 women) from the SPVM and other police forcesvolunteered for the study. Their mean age was 32.6 years (standard deviation = 7.7), their mean

number of years of schooling or education was 15.6 (standard deviation = 1.9), and 54% of themwere married or in a common-law relationship. All officers were currently in active service, andthe mean number of years of policing experience was 8.6. Table 1 gives the sociodemographiccharacteristics of the officers who took part in and completed the study.


25/81


Table 1 – Sociodemographic characteristics of police officers exposed to a traumatic event

( N = 83)

Sociodemographic data % M (SD)

GenderMale (n = 63)

Female (n = 20)

76

24

Age at time of study ( N = 83) 32.6 (7.7)

Marital status

Single (n = 32)

In relationship (n = 45)

Separated or divorced (n = 6)

39

54

7

Number of children

One (n = 9)

Two (n = 16)

Three or more (n = 9)

11

19

11

Number of years of schooling or education ( N = 83) 15.6 (1.9)

Ethnic origin

Caucasian

Ethnic minority

Afro-Canadian

AsianHispanic

92

8

5

21

Number of years of experience ( N = 83) 8.6 (7.3)

Number of hours worked per week ( N = 83) 35.4 (3.2)

Note: M = mean; SD = standard deviation

3.2 Measurement Instruments Used

Measurement instruments were chosen for their psychometric and clinical properties and becausethey were suited to ensure appropriate verification of the research findings and to address themethodological shortcomings of earlier studies. The vast majority of these instruments have beenvalidated in English and French. The French versions of the instruments were used, and theinterviews were conducted in French.


26/81


3.2.1 Diagnostic Measurement of PTSD and Other Mental Disorders

The modules of the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) [119]were used to assess the presence or absence of current or past mental disorders. The followingdisorders were evaluated: anxiety disorders, including acute stress disorder (ASD) and PTSD,

substance- or alcohol-abuse disorders, and major depressive disorder.For PTSD, full or partial diagnosis was assessed. A diagnosis of partial PTSD was assigned tosubjects who did not have sufficient symptoms to satisfy the conditions for full PTSD, but didhave at least one symptom under criterion B, C or D, or at least one symptom under criterion Band two symptoms under criterion D, as set out in DSM-IV [7]. This procedure for arriving at adiagnosis of partial PTSD follows the method used by Schnurr [36]. A diagnosis of subclinicalPTSD was also evaluated. This diagnosis was attributed when a participant did not meet theconditions for full or partial PTSD, but had at least one symptom that is listed under criterion B,C or D of DSM-IV [7].

The assessments were administered by trained, qualified assessors. The SCID-I was chosen because it is widely used, and its validity is well recognized for the diagnosis of psychiatricdisorders. Most studies of the psychometric properties of the SCID-I were conducted using the previous version (SCID-I, DSM-III-R). Nonetheless, the properties of the older version alsoapply to the current one (SCID-I, DSM-IV), as the modules are virtually identical [120]. Theearlier version had good concomitant validity with clinician judgment (κ = 0.69) [121], as wellas good convergent validity with other PTSD measurements, such as the Impact of Event Scale [122]. For inter-rater reliability, kappa values of 0.87 [112] and 0.77 [123] have been reported.The few studies that have been done with the current version of the SCID-I have reportedreliability coefficients comparable to, if not greater than, those of the earlier version [120]. Forthis study, independent assessors came to inter-rater agreements on a certain number of randomlyselected SCID-I interviews (ASD and PTSD module) at each measurement time: time 1 (44%; n= 35), time 2 (44%; n = 36), time 3 (49%; n = 39) and time 4 (50%, n = 38). Perfect agreement(κ = 1.0) was seen for the diagnoses of PTSD and ASD.

3.2.2 Measurement of Pretraumatic Factors

An ad hoc questionnaire was used to gather sociodemographic data (i.e., gender, age, level ofeducation, ethnic origin, marital status and number of children) and information about work-related characteristics (i.e., rank, number of hours of work per week and number of years of police experience) for each officer at the time of the TE. The questionnaire also served to assessan officer’s use of medical and psychological services in the 3 months prior to the TE, and thenumber of stressful events and their intensity (e.g., moving, legal problems, financial problems)experienced in the 12 months leading up to the TE.

A number of SCID-I modules can be used to evaluate whether a person was suffering fromPTSD or other medical disorders before the TE. Family psychological history was assessed bymeans of the following question: “Have any members of your family ever had psychological problems?” If so, participants were asked to indicate what their relationship was with that personand specify what kind of problems the person had.


27/81


The Coping Inventory for Stressful Situations (CISS) [124] is used to measure how people reactto situations they perceive as threatening. It is a 48-item measure of stress management abilitythat refers to three main coping styles: emotion-oriented, task-oriented and avoidance. The itemsare self-assessed on a five-point Likert scale where 1 signifies “not at all,” and 5 “very much.” Astudy of the French version of the instrument has shown solid internal validity [125]. The

coefficients for the Emotion (0.86), Task (0.87) and Avoidance (0.83) scales indicate that thethree dimensions are homogeneous. At the same time, the three scales are relatively independent[125].

The Life Events Checklist , part of the Clinician-Administered PTSD Scale (CAPS ; [126]), is a 17-item self-report measure designed to assess traumatic events that may have occurred inchildhood or adulthood (e.g., combat-related experiences, accidents, disasters, serious illness,sexual abuse). Respondents indicate whether they have experienced, witnessed or learned abouteach event. We also asked each participant how each event affected them psychologically, whichallowed us to distinguish merely stressful events (i.e., an event that meets only diagnosticcriterion A1 for PTSD, as defined in DSM-IV) from true traumatic events (i.e., an event thatmeets PTSD criteria A1 and A2 in DSM-IV). The French version of the questionnaire has beenvalidated with Quebec students [126]. The instrument possesses good internal consistency, witha coefficient of 0.89 and test-retest reliability of 0.97.

The French version [127] of the 15-Item Dispositional Resilience (Hardiness) Scale ( DRS ; [128])is a self-report scale designed to measure a person’s capacity to withstand stress. The instrumentconsists of 15 items spread equally among three subscales, corresponding to the three dimensionsof a resilient personality: commitment, control and challenge. The items are rated on a four-pointLikert scale where 0 = not at all true, and 3 = completely true. The English version possesses analpha coefficient of 0.84 for the overall score and 0.71 to 0.78 for the subscales [128]. TheFrench version of the DRS was validated with a French-speaking adult sampling from Quebec[127]. The internal consistency reliability coefficient is 0.66 for the overall assessment, while the

corresponding coefficients for the commitment, control and challenge subscales are 0.48, 0.48and 0.69. Test-retest reliability is 0.71.

The World Assumptions Scale [129] is a self-report questionnaire that measures individuals’fundamental beliefs about the world and their personal values. It contains 32 items, which arerated on a six-point Likert scale, with 1 meaning “strongly disagree” and 6 “strongly agree.” Novalidation study has been done of the original version of the questionnaire, nor of the translatedFrench version. On the other hand, analyses of the French version conducted with a clinicalsample of 30 victims and 35 people from the general community have shown that the instrumenthas good psychometric properties [130].

The Self-Efficacy Scale [131] is a self-report questionnaire that measures one’s general efficacyexpectations unrelated to specific situations or behaviour. It consists of 30 items, but 7 of themare not used in computing the total score. The basic premise underlying the development of thisinstrument is the observation that personal expectations about control are major determinants of behavioural change, and individual differences in past experiences and attribution of success produce distinct degrees of generalized self-efficacy expectations. This instrument can therefore be useful in assessing the process of adaptation to the client’s needs over the course of clinicaltreatment. It is also an indicator of progress, as expectations about self-efficacy should change


28/81


over the course of treatment. The scale has two self-efficacy subscales: a general one (17 items)and a social one (6 items). The instrument has good internal consistency, with an alpha of 0.86for the general subscale, and 0.71 for the social subscale. No test-retest reliability has beenreported. The instrument also has good criterion validity, as well as good construct validity.

3.2.3 Measurement of Peritraumatic Factors

Traumatic event severity was assessed by means of an ad hoc interview that takes into accountthe objective and subjective characteristics of the event, such as the degree of involvement in theevent (i.e., direct involvement, direct or indirect witnessing); whether or not someone wasinjured or died as a result of the subject’s involvement, the duration of the event; itsuncontrollability, unforeseeable or unexpected nature; whether one’s life or physical integritywas threatened; whether there was verbal aggression, physical aggression, pain or physicalinjury; contact with the injured or the dead; use of firearms or other kinds of weapons during theevent. Points were assigned for each characteristic measured. The total number of items varied

between 4 and 32. A higher score reflects greater severity of exposure to the event.The Peritraumatic Dissociative Experiences Questionnaire–Self-Report Version (PDEQ-SRV) [132] is a 10-item instrument that assesses dissociative reactions experienced during andimmediately after a traumatic event. The items evaluate depersonalization, derealization,amnesia, out-of-body experience, change in perception of time and body image. Items are ratedon a five-point Likert scale, where 1 signifies “not at all true,” and 5 “extremely true.” The totalscore ranges from 10 to 50. This instrument has good reliability and validity coefficients [133]. Italso has good convergent, discriminant and predictive validity [134]. The French version of thequestionnaire was validated with a student population [135] and showed the same psychometric properties as the English version. It thus possesses good internal consistency, with a coefficientof 0.85 and test-retest reliability of 0.88.

The effect of fright was assessed when the participant answered yes to one or more of thefollowing questions:

• Were you stupefied or in a state of shock to the point where you couldn’t feel anything?• Did you, for a moment, experience a complete absence of emotion?•

Were you unable to think or to speak?• Did you feel like you were frozen?

The Initial Subjective Reaction Emotional Scale of the Potential Stressful Events Interview [136]is a self-report instrument for assessing emotional responses during and immediately after atraumatic event (e.g., fear, confusion, sadness, shame, surprise, anger). The 15 items are rated on

a four-point Likert scale, where 1 means “not at all,” and 4 “completely.” The total item scoreranges from 15 to 60. A principal component analysis has shown that 4 of the 15 items aresufficiently reliable, corresponding to the fear, dissociation, guilt and anger subscales. Internalconsistency coefficients vary between 0.62 and 0.79. The convergent validity, discriminantvalidity and reliability of the questionnaire are satisfactory [136].

The Initial Subjective Reaction Physical Scale of the Potential Stressful Events Interview [136] isa 10-item self-report scale that assesses peritraumatic physical reactions at the time of a


29/81


traumatic event. It measures 10 symptoms associated with panic attacks, such as sweating, breathlessness, dizziness, heart palpitations and hot flushes. Items are rated on a four-point Likertscale, where 1 signifies “not at all,” and 4 “completely.” The total score ranges from 10 to 40.The instrument has good internal consistency, with a coefficient of 0.86 [59].

Two questions were developed by the research team to assess the perception of the quality of the positive support received from coworkers and superiors during or immediately after thetraumatic event. The first question read: “In your opinion, did your coworkers offer you adequatesupport during or immediately after the event?” The second question was worded the same way, but assessed the perception of the positive support received from superiors or supervisors. Itemsare rated on a five-point Likert scale, where 1 means “not at all,” and 5 “completely.”

The research team also developed some specific questions ad hoc to assess confidence in one’sability to act effectively during the event , confidence in having been trained adequately to dealwith such an event , confidence in having sufficient experience as a police officer to deal withsuch an event and the level of control perceived during the event .

3.2.4

Measurement of Posttraumatic Factors

Acute stress disorder (ASD) was assessed in the first month following the traumatic event usingthe SCID-I .

The Modified PTSD Symptom Scale (MPSS) [137] is used to assess the frequency and severity of posttraumatic symptoms experienced over the last two weeks. It consists of 17 items thatmeasure PTSD symptoms on two Likert scales, one for frequency (number of times per week)and the other for severity (level of disturbance). The frequency items are rated on a four-pointLikert scale (0 = not at all; 1 = once per week or less/a little/once in a while; 2 = 2 to 4 times perweek/some/half the time; 3 = 5 or more times per week/a lot/almost always). Responses for the

severity scale are on a five-point Likert scale (0 = not at all disturbing, to 4 = extremelydisturbing). The overall score combines the sum of the scores obtained on the frequency andseverity scales. The total score can therefore range from 0 to 119. The English version hasinternal consistency of 0.91, specificity of 83.8% and concurrent validity of 0.92 with the SCID.Validated with a Quebec clinical sample [138], this instrument possesses excellent internalconsistency: 0.92 for the frequency scale, and 0.95 for the severity scale. Temporal stability is0.98 for the two scales.

The Perceived Support Inventory (PSI) [139] is a self-report consisting of two subscales:(1) Perceived Supportive Spouse Behaviours [140] measures an individual’s perception of thefrequency of positive social support received from the person’s most significant other. It contains

11 items divided between 2 factors to measure tangible and emotional support, both general andspecific to the problems experienced by the participant. The Quebec version possesses very goodinternal consistency (0.87) and a coefficient of convergent validity with the Social ProvisionsScale of r = 0.44 [141]; (2) Perceived Negative Spouse Behaviours [140] measures a person’s perception of the frequency of negative behaviours and social interactions from his or her mostsignificant other. It has 13 items split between 2 factors to measure either withdrawal/avoidanceor criticism. The Quebec version has very good internal consistency (0.84) and a coefficient ofconvergent validity with the Social Provisions Scale of r = -0.43 [141]. For the two subscales of


30/81


the Perceived Support Inventory, perceptions of the frequency of behaviour are measured using afour-point Likert scale ranging from 1 (never responded that way) to 4 (often responded thatway). Respondents are asked to indicate their perception of supportive behaviours since thetraumatic event or since the last time an assessment was done.

Our own questionnaire Soutien organisationnel [Organizational Support] was used to gaugetangible assistance to police officers after the event. When answering the questionnaire, officershad to indicate whether, following the traumatic event or since the last measurement was done,they had benefited from paid leave, arrangements or changes in their assigned duties,compensation, consultations with a psychologist, consultations with a member of the police brotherhood, consultations with a police resource person or psychological debriefing. Thequestionnaire also assessed officers’ sense of appreciation for the work they do, their feeling of being effective in their work and to what extent media coverage of the event had affected them.The questionnaire was administered at all assessment times.

The Group Environment Scale [142] serves to measure three specific dimensions of the socialenvironment: relationships, personal growth, and the maintenance and change system in thegroup. The 90 items on the questionnaire are divided into 10 subscales, with three dimensions.The scale possesses acceptable internal consistency, test-retest reliability ranging from 0.67 to0.87 for the subscales and excellent stability over time. Its construct validity is also good. For the purposes of this study, we used only three of the subscales (group cohesion, leader support, andexpressiveness), which are all part of the relationship dimension. Each subscale has 9 items,which take the form of true or false statements. The total score for each subscale can thereforerange from 0 to 9, where a higher score represents, respectively, better group cohesion, betterleadership or better expressiveness on the part of the group members.

The Beck Depression Inventory-II (BDI) [143] determines whether or not a person has hadsymptoms of depression over the last two weeks, and if so, how severe they are. It is a 21-itemself-report questionnaire. Each question is answered on a four-point Likert scale, from 0 to 3.The total overall score can therefore range from 0 to 63. The instrument has been validated forQuebec women and men [144]. The internal consistency coefficient is 0.82. Rest-retest reliabilityfor a three-month interval is 0.75. Construct validity is also good. A psychometric study ofFrench-speaking university students [145] revealed good concurrent validity when the instrumentis compared with other instruments for measuring depression. The questionnaire wasadministered at all assessment times.

The ad hoc questionnaire on use of medical and psychological services [Recours aux servicesmédicaux et psychologiques] documents the use of health care services before and after theevent, in an effort to determine the impact of a traumatic event on participants’ mental and

physical health. All consultations with general practitioners, medical specialists, psychologists, psychiatrists or alternative medical practitioners were recorded, regardless whether or not theconsultation had to do with TE-related symptoms or problems. The questionnaire wasadministered at T1 to measure the use of medical and psychological services in the three months before the traumatic event, at T3 to measure services received in the three months following theTE and at T4 to measure care received in the last three months. The existence or absence of painand physical injuries stemming from the TE was documented, as well as officers’ perception of


31/81


the severity of their pain/injuries. Participants also indicated whether or not they had receivedmedical services in connection with their pain/injuries.

Table 2 provides an overview of the different variables measured for the study.


32/81


33/81


Hardiness (stress-resistant personality) French version of Dispositional Resilience(Hardiness) Scale (DRS) [128]

X

Beliefs about the world and personal values World Assumptions Scale [129] X X X X

Sense of self-efficacy Self-efficacy Scale [131] X X X X

Peritraumatic variables

Objective and subjective severity of TE Ad hoc questionnaire X

Initial dissociative reactions Peritraumatic Dissociative ExperiencesQuestionnaire – Self-Report Version (PDEQ-SRV)[132]

X

Fright Ad hoc questions X

Initial emotional reactions Initial Subjective Reaction Emotional Scale of thePotential Stressful Events Interview [136]

X

Initial physical reactions Initial Subjective Reaction Physical Scale of thePotential Stressful Events Interview [136]

X

Perception of support from coworkers and

superiors

Ad hoc question X

Confidence in ability to intervene effectively Ad hoc question XConfidence in training to deal with event Ad hoc question XConfidence in having sufficient experience as a police officer to deal with event

Ad hoc question X

Perceived level of control during event Ad hoc question X

Posttraumatic variablesAcute stress disorder Structured Clinical Interview for DSM-IV (SCID-I)

[119] X X


34/81


PTSD diagnosis (clinical and partial) Structured Clinical Interview for DSM-IV (SCID-I)[119]

X X X

Frequency and severity of PTSD symptoms Modified PTSD Symptom Scale (MPSS)[137] X X X

Perception of positive social support fromsignificant others

Perceived Supportive Spouse Behaviours of thePerceived Support Inventory (PSI) [139]

X X X X

Perception of negative social interactions with

significant others

Perceived Negative Spouse Behaviours of the

Perceived Support Inventory (PSI) [139]

X X X X

Organizational support received Ad hoc questionnaire X X X X

Social climate of group Group Environment Scale [142] X X X X

Symptoms of depression Beck Depression Inventory-II (BDI) [143] X X X XUse of medical and psychological services in last3 months

Ad hoc questionnaire X X

Number of stressful events experienced since lasttime of assessment (e.g., moving, legal problems, financial problems, etc.)

Life Stress Event Scale X X X


35/81


3.3

Study Design and Variables

This study was based on a repeated-measures prospective research protocol and therefore

followed a quasi-experimental design. The dependent variables—the diagnosis of PTSD (clinical

and partial) and the severity of symptoms—were measured at each assessment time. The

independent variables consisted of the assessment time, and the pretraumatic, peritraumatic and posttraumatic risk and protective factors. The pretraumatic factors examined were

sociodemographic characteristics, police experience, age at the time of the TE, coping strategies,hardiness, fundamental beliefs, history of personal and family mental problems, and previous

stressful and traumatic events. The peritraumatic factors were the severity of the event, fright,

initial emotional and physical stress reactions, dissociative reactions and the perception of thesupportiveness of coworkers and superiors during or immediately after the event. The

posttraumatic factors studied were the number of ASD symptoms, symptoms of depression, and

perception of social support (i.e., positive support and negative social interactions, organizationalsupport, group environment, use of medical and psychological services).

Table 3 – Research protocol for repeated-measures prospective study

Variables

Assessment times

T1

(5 to 15 days after

event)

T2

(1 month)

T3

(3 months)

T4

(12 months)

Independentvariables

X X X X

Dependent

variables

X X X

3.4

Sample Size

The sample size required for this prospective study was established by taking into account the

main statistical analyses, i.e., multivariate linear regression, planned for testing our researchhypotheses. To determine the sample size required to test the null hypothesis of the relative risk

in a prospective study, we used the procedure described by Lemeshow, Hosmer, Klar and

Lwanga [146]. For type I errors of 0.05, type II errors of 0.20, and a minimum effect size of p =

0.5, the one-sided test required a sample of 94 police officers exposed to a critical incident. Since

we estimated we would face an attrition rate of approximately 10%, we planned to

predictores de desroden de estres

Documents

Transcript of predictores de desroden de estres